Prior work by our group (Spiegel, et al, 1989) has provided evidence that participation of metastatic breast cancer patients in group treatment prolongs survival time am average of eighteen months. This difference is clinically as well as statistically significant. This intervention is proposed as a model system for examining psychosocial effects on the progression of a terminal illness. The prognosis and course of metastatic breast cancer and HIV/AIDS are rather similar. Both are almost uniformly fatal, but have a staged and predictable course, usually 2-3 years from diagnosis of breast metastases or AIDS to death. There are four specific aims of the study. We propose to examine: (1) The medical effects of psychosocial treatment for patients with metastatic carcinoma. We predict enhanced survival time among the patients randomized to group support. (2) The medical basis of prolonged survival in cancer patients. We will test the hypothesis that as a result of decreased anxiety, depression and pain, patients who participate in group treatment increase exercise, improve their diet, and undertake more vigorous medical treatment, thereby, enhancing survival. In addition, we will study other physiological mechanisms of delayed disease progression, including the stress hormone cortisol and in vivo immune measure, delayed hypersensitivity. Treatment effects could also be mediated via reduced cortisol or enhanced immune function. Changes in these measures would be relevant to an HIV/AIDS population. (3) psychosocial effects of group treatment. We plan to study the effects of participation in the group on mood disturbance using specific measures of anxiety and depression, pain, and family environment. We predict reduced anxiety, depression, and pain in the treatment group. (4) The predictive power of personality variables, especially repression of emotion, mood, and family environment, in determining survival time. The design of the study involves the recruitment of a well-defined sample of 100 metastatic breast cancer patients. These patients will be matched to assure comparability of medical condition and randomized to a year of group treatment or to routine oncological care. This study will provide an empirical basis for fundamentally altering the structure of standard oncological care for metastatic breast cancer, and, as a model for other patients with progressive terminal illnesses, such as AIDS. Those factors that are found to be determinative of increased survival time will provide a rationale for designing new treatment programs. These studies will determine whether psychosocial support should be added to the standard treatment regimen because of proven medical as well as psychological benefit.